U.S. patent number 10,340,855 [Application Number 15/772,133] was granted by the patent office on 2019-07-02 for doherty amplifier.
This patent grant is currently assigned to Mitsubishi Electric Corporation. The grantee listed for this patent is Mitsubishi Electric Corporation. Invention is credited to Yuji Komatsuzaki, Keigo Nakatani, Shintaro Shinjo, Takaaki Yoshioka.
View All Diagrams
United States Patent |
10,340,855 |
Komatsuzaki , et
al. |
July 2, 2019 |
Doherty amplifier
Abstract
A Wilkinson power divider includes: .pi.-type LPFs connected to
an input terminal; a T-type HPF having one end connected to one of
the .pi.-type LPFs and having another end connected to a carrier
amplifier; another T-type HPF having one end connected to another
one of the .pi.-type LPFs and having another end connected to a
.lamda./4 line; and an isolation resistor connected to connection
points.
Inventors: |
Komatsuzaki; Yuji (Tokyo,
JP), Shinjo; Shintaro (Tokyo, JP),
Nakatani; Keigo (Tokyo, JP), Yoshioka; Takaaki
(Tokyo, JP) |
Applicant: |
Name |
City |
State |
Country |
Type |
Mitsubishi Electric Corporation |
Tokyo |
N/A |
JP |
|
|
Assignee: |
Mitsubishi Electric Corporation
(Tokyo, JP)
|
Family
ID: |
59274513 |
Appl.
No.: |
15/772,133 |
Filed: |
January 5, 2016 |
PCT
Filed: |
January 05, 2016 |
PCT No.: |
PCT/JP2016/050095 |
371(c)(1),(2),(4) Date: |
April 30, 2018 |
PCT
Pub. No.: |
WO2017/119062 |
PCT
Pub. Date: |
July 13, 2017 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20180287566 A1 |
Oct 4, 2018 |
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H03F
3/189 (20130101); H03F 1/56 (20130101); H03F
1/0288 (20130101); H03H 7/0115 (20130101); H03H
7/07 (20130101); H03F 1/07 (20130101); H03H
7/1783 (20130101); H03F 3/19 (20130101); H03F
2200/255 (20130101); H03F 2200/267 (20130101); H03F
2200/423 (20130101); H03H 7/48 (20130101); H03F
2200/451 (20130101); H03F 2200/165 (20130101) |
Current International
Class: |
H03F
1/07 (20060101); H03F 3/19 (20060101); H03H
7/07 (20060101); H03H 7/01 (20060101); H03F
3/189 (20060101); H03F 1/02 (20060101); H03F
1/56 (20060101); H03H 7/48 (20060101) |
Field of
Search: |
;330/124R,295,302 |
Foreign Patent Documents
|
|
|
|
|
|
|
2006-333022 |
|
Dec 2006 |
|
JP |
|
2006-339981 |
|
Dec 2006 |
|
JP |
|
2011-114492 |
|
Jun 2011 |
|
JP |
|
Primary Examiner: Choe; Henry
Attorney, Agent or Firm: Studebaker & Brackett PC
Claims
The invention claimed is:
1. A Doherty amplifier comprising: division circuitry configured to
split, between a first transmission line and a second transmission
line, a signal to be amplified; first amplifier circuitry inserted
into the first transmission line; second amplifier circuitry
inserted into the second transmission line; and a power combiner
configured to combine signals amplified by the first and second
amplifier circuitry, wherein the division circuitry includes a
first filter to which the signal to be amplified is input, a second
filter connected between the first filter and the first amplifier
circuitry, a third filter to which the signal to be amplified is
input, a fourth filter connected between the third filter and the
second amplifier circuitry, and a resistor connected to an output
side of the first filter and an output side of the third filter,
wherein each of the first and third filters is a low-pass filter
while each of the second and fourth filters is a high-pass filter,
or wherein each of the first and third filters is a high-pass
filter while each of the second and fourth filters is a low-pass
filter, and wherein, when each of the low-pass filters is formed by
a .pi.-type circuit, each of the high-pass filters is formed by a
T-type circuit, and when each of the low-pass filters is formed by
a T-type circuit, each of the high-pass filters is formed by a
.pi.-type circuit.
2. The Doherty amplifier according to claim 1, wherein the first
amplifier circuitry includes a carrier amplifier connected to an
output side of the second filter, and a first n-quarter wavelength
line connected between the carrier amplifier and the power
combiner, the first n-quarter wavelength line having an electrical
length being an n-quarter (n is a positive odd number) wavelength
of the signal to be amplified, and the second amplifier circuitry
includes a second n-quarter wavelength line having one end
connected to an output side of the fourth filter and having an
electrical length being an n-quarter (n is a positive odd number)
wavelength of the signal to be amplified, and a peak amplifier
connected between another end of the second n- quarter wavelength
line and the power combiner.
3. The Doherty amplifier according to claim 2, wherein each of the
first and third filters is the low-pass filter of the .pi.-type
circuit while each of the second and fourth filters is the
high-pass filter of the T-type circuit, and the second n-quarter
wavelength line is formed by a low-pass filter with a .pi.-type
circuit.
4. The Doherty amplifier according to claim 2, wherein each of the
first and third filters is the high-pass filter of the .pi.-type
circuit while each of the second and fourth filters is the low-pass
filter of the T-type circuit, and the second n-quarter wavelength
line is formed by a high-pass filter with a .pi.-type circuit.
5. The Doherty amplifier according to claim 2, wherein each of the
first and third filters is the high-pass filter of the T-type
circuit while each of the second and fourth filters is the low-pass
filter of the .pi.-type circuit, and the second n-quarter
wavelength line is formed by a high-pass filter with a T-type
circuit.
6. The Doherty amplifier according to claim 2, wherein each of the
first and third filters is the low-pass filter of the T-type
circuit while each of the second and fourth filters is the
high-pass filter of the .pi.-type circuit, and the second n-quarter
wavelength line is formed by a low-pass filter with a T-type
circuit.
7. The Doherty amplifier according to claim 1, wherein the first
amplifier circuitry includes a first n-quarter wavelength line
having one end connected to an output side of the second filter and
having an electrical length being an n-quarter (n is a positive odd
number) wavelength of the signal to be amplified, and a carrier
amplifier connected between another end of the first n- quarter
wavelength line and the power combiner, and the second amplifier
circuitry includes a peak amplifier connected to an output side of
the fourth filter, and a second n-quarter wavelength line connected
between the peak amplifier and the power combiner, the second
n-quarter wavelength line having an electrical length being an
n-quarter (n is a positive odd number) wavelength of the signal to
be amplified.
8. The Doherty amplifier according to claim 7, wherein each of the
first and third filters is the low-pass filter of the .pi.-type
circuit while each of the second and fourth filters is the
high-pass filter of the T-type circuit, and the first n-quarter
wavelength line is formed by a low-pass filter with a .pi.-type
circuit.
9. The Doherty amplifier according to claim 7, wherein each of the
first and third filters is the high-pass filter of the .pi.-type
circuit while each of the second and fourth filters is the low-pass
filter of the T-type circuit, and the first n-quarter wavelength
line is formed by a high-pass filter with a .pi.-type circuit.
10. The Doherty amplifier according to claim 7, wherein each of the
first and third filters is the high-pass filter of the T-type
circuit while each of the second and fourth filters is the low-pass
filter of the .pi.-type circuit, and the first n-quarter wavelength
line is formed by a high-pass filter with a T-type circuit.
11. The Doherty amplifier according to claim 7, wherein each of the
first and third filters is the low-pass filter of the T-type
circuit while each of the second and fourth filters is the
high-pass filter of the .pi.-type circuit, and the first n-quarter
wavelength line is formed by a low-pass filter with a T-type
circuit.
Description
TECHNICAL FIELD
The present invention relates to a Doherty amplifier capable of
amplifying a signal over a wide band.
BACKGROUND ART
In recent years, Doherty amplifiers have been proposed as high
efficiency amplifiers for communications.
In a Doherty amplifier, a carrier amplifier being biased as class
AB or class B and a peak amplifier being biased as class C are
connected in parallel.
Division circuitry is connected at a preceding stage of the carrier
amplifier and the peak amplifier, which are connected in parallel.
The division circuitry splits a signal to be amplified between the
carrier amplifier and the peak amplifier.
Further, a power combiner for combining the signals amplified by
the carrier amplifier and the peak amplifier is connected at a
subsequent stage of the carrier amplifier and the peak
amplifier.
Note that the carrier amplifier performs signal amplification
continuously, whereas the peak amplifier performs signal
amplification only when high power output is required.
A Doherty amplifier disclosed in Patent Literature 1 below
comprises division circuitry and an isolation resistor. The
division circuitry includes four .lamda./4 lines having
characteristic impedance that depends on a sprit ratio of signal
power to a carrier amplifier and a peak amplifier. The isolation
resistor has a resistance value that depends on the sprit ratio of
the signal power.
The .lamda./4 line is a distributed constant line having an
electrical length that is a quarter wavelength of a signal to be
amplified.
CITATION LIST
Patent Literature 1: JP 2006-339981 A (for example, FIG. 1)
SUMMARY OF INVENTION
Since the conventional Doherty amplifier has the structure
explained above, a signal loop may occur from the division
circuitry to return thereto through the carrier amplifier, the
power combiner, and the peak amplifier. In a case where this loop
has a gain, loop oscillation occurs. As a result, there arises a
problem that a stabilizing circuit for suppressing the loop
oscillation needs to be mounted, leading to an increase in size of
the circuit.
If increasing the resistance value of the isolation resistor
mounted in the division circuitry, the loop oscillation can be
suppressed without mounting the stabilizing circuit. However, there
is a problem that a passage characteristic of a desired band
becomes narrow due to the increase in the resistance value of the
isolation resistor, and thereby the signal cannot be amplified over
a wide band.
The present invention has been made to solve the above problems,
and an objective thereof is to provide a Doherty amplifier which is
capable of suppressing loop oscillation without mounting a
stabilizing circuit and also capable of amplifying a signal over a
wide band.
A Doherty amplifier according to the present invention includes:
division circuitry configured to split, between a first
transmission line and a second transmission line, a signal to be
amplified; first amplifier circuitry inserted into the first
transmission line; second amplifier circuitry inserted into the
second transmission line; and a power combiner configured to
combine signals amplified by the first and second amplifier
circuitry, wherein the division circuitry includes a first filter
to which the signal to be amplified is input, a second filter
connected between the first filter and the first amplifier
circuitry, a third filter to which the signal to be amplified is
input, a fourth filter connected between the third filter and the
second amplifier circuitry, and a resistor connected to an output
side of the first filter and an output side of the third filter,
wherein each of the first and third filters is a low-pass filter
while each of the second and fourth filters is a high-pass filter,
or wherein each of the first and third filters is a high-pass
filter while each of the second and fourth filters is a low-pass
filter, and wherein, when each of the low-pass filters is formed by
a .pi.-type circuit, each of the high-pass filters is formed by a
T-type circuit, and when each of the low-pass filters is formed by
a T-type circuit, each of the high-pass filters is formed by a
.pi.-type circuit.
According to the present invention, the division circuitry includes
the first filter to which the signal to be amplified is input, the
second filter connected between the first filter and the first
amplifier circuitry, the third filter to which the signal to be
amplified is input, the fourth filter connected between the third
filter and the second amplifier circuitry, and the resistor
connected to an output side of the first filter and an output side
of the third filter, wherein each of the first and third filters is
a low-pass filter while each of the second and fourth filters is a
high-pass filter, or wherein each of the first and third filters is
a high-pass filter while each of the second and fourth filters is a
low-pass filter, and wherein, when each of the low-pass filters is
formed by a .pi.-type circuit, each of the high-pass filters is
formed by a T-type circuit, and when each of the low-pass filters
is formed by a T-type circuit, each of the high-pass filters is
formed by a .pi.-type circuit. Therefore, it is capable of bringing
advantages that loop oscillation is suppressed without mounting a
stabilizing circuit and a signal is amplified over a wide band.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 1 of the present invention.
FIG. 2 is a Smith chart illustrating frequency characteristics of
impedance in .pi.-type LPFs 2a and 2c and T-type HPFs 2b and
2d.
FIG. 3A is a configuration diagram illustrating a circuit structure
of a .pi.-type LPF, and FIG. 3B is a configuration diagram
illustrating a circuit structure of a T-type LPF.
FIG. 4A is a configuration diagram illustrating a circuit structure
of a .pi.-type HPF, and FIG. 4B is a configuration diagram
illustrating a circuit structure of a T-type HPF.
FIG. 5 is an explanatory diagram illustrating an isolation
characteristic of a Wilkinson power divider 2.
FIG. 6A is an explanatory diagram illustrating a passage
characteristic of a Wilkinson power divider 2, FIG. 6B is an
explanatory diagram illustrating frequency characteristics of
impedance in a .pi.-type LPF and a .pi.-type HPF, and FIG. 6C is an
explanatory diagram illustrating frequency characteristics of
impedance in a .pi.-type LPF and a T-type HPF.
FIG. 7 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 2 of the present invention.
FIG. 8 is a Smith chart illustrating frequency characteristics of
impedance in .pi.-type HPFs 40a and 40c and T-type LPFs 40b and
40d.
FIG. 9 is an explanatory diagram illustrating an isolation
characteristic of a Wilkinson power divider 40.
FIG. 10A is an explanatory diagram illustrating a passage
characteristic of the Wilkinson power divider 40, FIG. 10B is an
explanatory diagram illustrating frequency characteristics of
impedance in a .pi.-type HPF and a .pi.-type LPF, and FIG. 10C is
an explanatory diagram illustrating frequency characteristics of
impedance in a .pi.-type HPF and a T-type LPF.
FIG. 11 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 3 of the present invention.
FIG. 12 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 4 of the present invention.
FIG. 13 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 5 of the present invention.
FIG. 14 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 6 of the present invention.
FIG. 15 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 7 of the present invention.
FIG. 16 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 8 of the present invention.
FIG. 17 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 9 of the present invention.
FIG. 18 is a configuration diagram illustrating a Doherty amplifier
according to the Embodiment 9 of the present invention.
FIG. 19 is a configuration diagram illustrating a Doherty amplifier
according to the Embodiment 9 of the present invention.
FIG. 20 is a configuration diagram illustrating a Doherty amplifier
according to the Embodiment 9 of the present invention.
FIG. 21 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 10 of the present invention.
FIG. 22 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 11 of the present invention.
FIG. 23 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 12 of the present invention.
FIG. 24 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 13 of the present invention.
DESCRIPTION OF EMBODIMENTS
Hereinafter, to describe the present invention in more detail,
embodiments for implementing the present invention will be
described with reference to the accompanying drawings.
Embodiment 1
FIG. 1 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 1 of the present invention.
In FIG. 1, an input terminal 1 is a terminal to which a signal to
be amplified is input.
A Wilkinson power divider 2 is division circuitry that splits the
signal to be amplified, which is input from the input terminal 1,
between a transmission line 3 as a first transmission line and a
transmission line 4 as a second transmission line.
The Wilkinson power divider 2 includes .pi.-type LPFs 2a and 2c,
each being a .pi.-type circuit low-pass filter, T-type HPFs 2b and
2d, each being a T-type circuit high-pass filter, and an isolation
resistor 2e.
The "LPF" is an abbreviation for "Low Pass Filter", and the "HPF"
is an abbreviation for "High Pass Filter".
The .pi.-type LPF 2a is a first filter having one end connected to
the input terminal 1.
The T-type HPF 2b is a second filter having one end connected to
the other end of the .pi.-type LPF 2a and having another end
connected to a carrier amplifier 6.
The .pi.-type LPF 2c is a third filter having one end connected to
the input terminal 1.
The T-type HPF 2d is a fourth filter having one end connected to
the other end of the .pi.-type LPF 2c and having another end
connected to one end of a .lamda./4 line 9.
The isolation resistor 2e is a resistor connected to a connection
point 2f and a connection point 2g. The connection point 2f
connects the .pi.-type LPF 2a and the T-type HPF 2b at an output
side of the .pi.-type LPF 2a. The connection point 2g connects the
.pi.-type LPF 2c and the T-type HPF 2d at an output side of the
.pi.-type LPF 2c. The isolation resistor 2e has a resistance value
that depends on a sprit ratio of signal power to the transmission
lines 3 and 4.
Amplifier circuitry 5 is first amplifier circuitry inserted into
the transmission line 3. The amplifier circuitry 5 includes the
carrier amplifier 6 and a .lamda./4 line 7.
The carrier amplifier 6 is connected to an output side of the
T-type HPF 2b in the Wilkinson power divider 2. The carrier
amplifier 6 amplifies one of signals split by the Wilkinson power
divider 2. Note that the carrier amplifier 6 is biased as class AB
or class B.
The .lamda./4 line 7a as a first n-quarter wavelength line is
connected between the carrier amplifier 6 and a power combiner 11.
The .lamda./4 line 7a is a distributed constant line that has an
electrical length being an n-quarter (n is a positive odd number)
wavelength of the signal to be amplified.
Amplifier circuitry 8 is second amplifier circuitry inserted into
the transmission line 4. The amplifier circuitry 8 includes the
.lamda./4 line 9 and a peak amplifier 10.
The .lamda./4 line 9 as a second n-quarter wavelength line is a
distributed constant line having one end connected to an output
side of the T-type HPF 2d in the Wilkinson power divider 2 and
having an electrical length being an n-quarter (n is a positive odd
number) wavelength of the signal to be amplified.
The peak amplifier 10 is connected between the .lamda./4 line 9 and
the power combiner 11 and amplifies the other signal which is split
by the Wilkinson power divider 2. Note that the peak amplifier 10
is biased as class C.
The power combiner 11 is a circuit that combines a signal amplified
by the amplifier circuitry 5 and a signal amplified by the
amplifier circuitry 8.
The output terminal 12 is a terminal for outputting a signal
combined by the power combiner 11.
The Doherty amplifier of FIG. 1 is designed to equalize electrical
lengths of one path and the other path at a desired band. The one
is a path from the input terminal 1 to an output terminal 12
through the .pi.-type LPF 2a, the T-type HPF 2b, the carrier
amplifier 6, the .lamda./4 line 7, and the power combiner 11. The
other is a path from the input terminal 1 to an output terminal 12
through the .pi.-type LPF 2c, the T-type HPF 2d, the .lamda./4 line
9, the peak amplifier 10, and the power combiner 11.
In the Embodiment 1, for simplifying description, the description
will be given on an assumption that the sprit ratio of the
Wilkinson power divider 2 is 1:1 and input/output impedance of the
Doherty amplifier is 50 .OMEGA..
FIG. 2 is a Smith chart illustrating frequency characteristics of
impedance in the .pi.-type LPFs 2a and 2c and the T-type HPFs 2b
and 2d.
As illustrated with a solid line in FIG. 2, the frequency
characteristic of impedance in the .pi.-type LPFs 2a and 2c, which
is obtained by viewing from the isolation resistor 2e, starts from
an arbitrary impedance that depends on the sprit ratio of the
Wilkinson power divider 2 at lower frequencies, and is minimized in
reflection at the desired band. That is, the frequency
characteristic is started from the impedance of 2.0 (100 .OMEGA.=50
.OMEGA..times.2.0) on a real axis of the Smith chart, and has the
minimum reflection in the desired band near the impedance of 1.0
(50 .OMEGA.=50 .OMEGA..times.1.0) on the real axis of the Smith
chart.
Thereafter, the frequency characteristic of impedance in the
.pi.-type LPFs 2a and 2c goes through a capacitive area as the
frequency becomes higher and eventually approaches a short point
where the impedance is 0 .OMEGA..
At this time, in the desired band, the impedance has a
characteristic to shift from lower resistance to higher resistance
along the real axis from lower frequencies to higher frequencies.
In FIG. 2, this characteristic is illustrated with the solid right
arrow.
FIGS. 3A and 3B are configuration diagrams illustrating circuit
structures of the LPFs. Specifically, FIG. 3A illustrates a circuit
structure of a .pi.-type LPF, and FIG. 3B illustrates a circuit
structure of a T-type LPF.
Each of the circuit structures of the .pi.-type LPFs 2a and 2c may
include an input terminal 21, an output terminal 22, capacitors 23
and 24, and an inductor 25, as illustrated in FIG. 3A.
However, each circuit structure for the .pi.-type LPFs 2a and 2c is
not limited to that in FIG. 3A as long as the circuit structure has
the frequency characteristic of impedance illustrated with the
solid line in FIG. 2. Therefore, the number of stages of each of
the .pi.-type LPFs 2a and 2c may be increased or decreased, and
each of the .pi.-type LPFs 2a and 2c may be formed by a distributed
constant line or the like.
As illustrated with a dash-dot line in FIG. 2, at lower
frequencies, the frequency characteristic of impedance in the
T-type HPFs 2b and 2d, which is obtained by viewing from the
isolation resistor 2e, starts from an open point, goes through a
capacitive area, and is minimized in reflection at the desired
band. That is, the frequency characteristic is started from
infinite impedance on the real axis of the Smith chart, and has the
minimum reflection in the desired band near the impedance of 1.0
(50 .OMEGA.=50 .OMEGA..times.1.0) on the real axis of the Smith
chart.
At higher frequencies, the frequency characteristic approaches
arbitrary impedance that depends on the sprit ratio of the
Wilkinson power divider 2. In the example of FIG. 2, the frequency
characteristic approaches the impedance of 50 .OMEGA..
At this time, in the desired band, the impedance has a
characteristic to shift from lower resistance to higher resistance
along the real axis from lower frequencies to higher frequencies.
In FIG. 2, this characteristic is illustrated with the dash-dot
right arrow.
FIGS. 4A and 4B are configuration diagrams illustrating circuit
structures of the HPFs. Specifically, FIG. 4A illustrates a circuit
structure of a i-type HPF, and FIG. 4B illustrates a circuit
structure of a T-type HPF.
The circuit structure of the T-type HPFs 2b and 2d may include an
input terminal 31, an output terminal 32, capacitors 36 and 37, and
an inductor 38, as illustrated in FIG. 4B.
However, each circuit structure for the T-type HPFs 2b and 2d is
not limited to that in FIG. 4B as long as the circuit structure has
the frequency characteristic of impedance illustrated with the
dash-dot line in FIG. 2. Therefore, the number of stages of each of
the T-type HPFs 2b and 2d may be increased or decreased, and each
of the T-type HPFs 2b and 2d may be formed by a distributed
constant line or the like.
Next, an operation will be described.
When a radio frequency (RF) signal of the desired band is input
from the input terminal 1 as the signal to be amplified, the
Wilkinson power divider 2 splits the RF signal between the
transmission line 3 and the transmission line 4.
The sprit ratio of the RF signal for the Wilkinson power divider 2
takes an arbitrary value depending on impedance transformations in
the .pi.-type LPFs 2a and 2c and the T-type HPFs 2b and 2d, and the
resistance value of the isolation resistor 2e.
One of the RF signals, which has been split by the Wilkinson power
divider 2 and output to the transmission line 3, is amplified by
the carrier amplifier 6. The other RF signal output to the
transmission line 4 is amplified by the peak amplifier 10.
The RF signal amplified by the carrier amplifier 6 and the other RF
signal amplified by the peak amplifier 10 are combined by the power
combiner 11. A resulting RF signal combined by the power combiner
11 is output through the output terminal 12.
The operation of the Wilkinson power divider 2 will be specifically
described.
FIG. 5 is an explanatory diagram illustrating an isolation
characteristic of the Wilkinson power divider 2.
In FIG. 5, in addition to the isolation characteristic of the
Wilkinson power divider 2 in FIG. 1, an isolation characteristic of
the conventional division circuitry disclosed in Patent Literature
1 is also depicted for comparison with the Wilkinson power divider
2 of FIG. 1. In the division circuitry disclosed in Patent
Literature 1, the four .lamda./4 lines are used in place of the
.pi.-type LPFs 2a and 2c and the T-type HPFs 2b and 2d.
In FIG. 5, in order to distinguish those characteristics, the
Wilkinson power divider 2 of FIG. 1 is labeled "present invention"
and the division circuitry disclosed in Patent Literature 1 is
labeled "conventional type".
As illustrated with the dotted lines in FIG. 5, the isolation
characteristic of the conventional-type division circuitry
indicates that isolation between the two output terminals in the
division circuitry becomes high at the vicinity of the desired
band, whereas it becomes low outside the desired band.
In contrast, as illustrated with the solid lines in FIG. 5, the
isolation characteristic of the Wilkinson power divider 2 of FIG. 1
indicates that isolation between the two output terminals in the
Wilkinson power divider 2 becomes high not only in the vicinity of
the desired band but also outside the desired band. The two output
terminals of the Wilkinson power divider 2 correspond to the output
side of the T-type HPF 2b connected to the carrier amplifier 6 and
the output side of the T-type HPF 2d connected to the .lamda./4
line 9.
The reason why the isolation becomes high even outside the desired
band as described above is that, while the T-type HPFs 2b and 2d
function to isolate a frequency band lower than the desired band,
the .pi.-type LPFs 2a and 2c function to isolate a frequency band
higher than the desired band.
FIGS. 6A to 6C are explanatory diagrams illustrating frequency
dependency of a passage characteristic in the Wilkinson power
divider 2 of FIG. 1.
Specifically, FIG. 6A illustrates a passage characteristic of the
Wilkinson power divider 2. FIG. 6B illustrates frequency
characteristics of impedance in a .pi.-type LPF and a .pi.-type
HPF. FIG. 6C illustrates frequency characteristics of impedance in
a .pi.-type LPF and a T-type HPF.
Note that, a circuit structure of the .pi.-type HPF may be formed
with the input terminal 31, the output terminal 32, inductors 33
and 34, and a capacitor 35 illustrated in FIG. 4A.
In FIG. 6B, the Wilkinson power divider has a circuit structure in
which an input-side filter is the .pi.-type LPF and an output-side
filter is the .pi.-type HPF. The .pi.-type LPF has a frequency
characteristic of impedance such that, the impedance shifts from
lower resistance to higher resistance along the real axis from
lower frequencies to higher frequencies in the desired band in the
vicinity of impedance of 1.0 (50 .OMEGA.=50 .OMEGA..times.1.0) on
the real axis of the Smith chart. In FIG. 6B, this characteristic
is illustrated with the solid right arrow.
In contrast, the .pi.-type HPF has a frequency characteristic of
impedance, such that the impedance shifts from higher resistance to
lower resistance along the real axis from lower frequencies to
higher frequencies in the desired band. In FIG. 6B, this
characteristic is illustrated with the dotted left arrow.
Therefore, the frequency characteristics of impedance in the
.pi.-type LPF and the .pi.-type HPF indicate that the individual
impedances shift in opposite directions along the real axis from
the lower frequencies to higher frequencies in the desired band. As
a result, only a center frequency of the desired band is matched,
and thereby a passband (S21) becomes a narrowband.
The passage characteristic (S21) illustrated with the dotted line
in FIG. 6A represents that of the Wilkinson power divider
illustrated in FIG. 6B. The passage characteristic (S21) indicates
a narrowband characteristic in which only a signal at the center
frequency of the desired band is allowed to pass through without
attenuation.
The Wilkinson power divider 2 of FIG. 1, which is illustrated in
FIG. 6C, has a circuit structure in which an input-side filter is
the .pi.-type LPFs 2a and 2c, and an output-side filter is the
T-type HPFs 2b and 2d. The .pi.-type LPFs 2a and 2c have a
frequency characteristic of impedance in which the impedance shifts
from lower resistance to higher resistance along the real axis from
lower frequencies to higher frequencies in the desired band. In
FIG. 6C, this characteristic is illustrated with the solid right
arrow.
In addition, the T-type HPFs 2b and 2d have a frequency
characteristic of impedance in which the impedance shifts from
lower resistance to higher resistance along the real axis from
lower frequencies to higher frequencies of the desired band. In
FIG. 6C, this characteristic is illustrated with the dash-dot right
arrow.
Therefore, the frequency characteristics of impedance in the
.pi.-type LPFs 2a and 2c and the T-type HPFs 2b and 2d indicate
that the impedance shifts in the same direction along the real axis
from lower frequencies to higher frequencies in the desired band.
As a result, frequencies are broadly matched with each other even
in the vicinity of the center frequency of the desired band, and
thereby the passband (S21) becomes a wide band.
The passage characteristic (S21) illustrated with the solid line in
FIG. 6A represents the passage characteristic of the Wilkinson
power divider 2 of FIG. 1 included in FIG. 6C. The passage
characteristic (S21) has a wide-band passage characteristic which
includes frequencies in the vicinity of the center frequency of the
desired band.
As is clear from the above, according to the Embodiment 1, the
Wilkinson power divider 2 includes: the .pi.-type LPFs 2a and 2c
connected to the input terminal 1; the T-type HPF 2b having one end
connected to an end of the .pi.-type LPF 2a and having another end
connected to the carrier amplifier 6; the T-type HPF 2d having one
end connected to an end of the .pi.-type LPF 2c and having another
end connected to an end of the .lamda./4 line 9, and the isolation
resistor 2e connected between the connection point 2f and the
connection point 2g. Therefore, it is capable of bringing
advantages that loop oscillation is suppressed without mounting a
stabilizing circuit and a signal is amplified over a wide band.
That is, by providing the Wilkinson power divider 2 achieving high
isolation, loop oscillation can be suppressed not only in the
vicinity of the desired band but also outside the desired band
without mounting a stabilizing circuit. In addition, a signal can
be amplified over a wide band because the .pi.-type LPFs 2a and 2c
and the T-type HPFs 2b and 2d have the frequency characteristics of
impedance, in which the impedance shifts in the same direction
along the real axis from the lower frequencies to higher
frequencies in the desired band.
Embodiment 2
In the foregoing Embodiment 1, the Wilkinson power divider 2, in
which input-side filters are the .pi.-type LPFs 2a and 2c while
output-side filters are the T-type HPFs 2b and 2d, splits the
signal to be amplified between the transmission lines 3 and 4.
Alternatively, a Wilkinson power divider, in which input-side
filters are .pi.-type HPFs while output-side filters are T-type
LPFs, may be used to split a signal to be amplified between
transmission lines 3 and 4.
FIG. 7 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 2 of the present invention. In FIG. 7,
the same reference signs as those in FIG. 1 denote the same or
corresponding part, and thus the description thereof will be
omitted.
A Wilkinson power divider 40 is division circuitry that splits a
signal to be amplified, which is input from an input terminal 1,
between the transmission line 3 and the transmission line 4.
The Wilkinson power divider 40 includes .pi.-type HPFs 40a and 40c,
each being a .pi.-type circuit high-pass filter, T-type LPFs 40b
and 40d, each being a T-type circuit low-pass filter, and an
isolation resistor 40e.
The .pi.-type HPF 40a is a first filter having one end connected to
the input terminal 1.
The T-type LPF 40b is a second filter having one end connected to
the other end of the .pi.-type HPF 40a and having another end
connected to a carrier amplifier 6.
The .pi.-type HPF 40c is a third filter having one end connected to
the input terminal 1.
The T-type LPF 40d is a fourth filter having one end connected to
the other end of the .pi.-type HPF 40c and having another end
connected to one end of a .lamda./4 line 9.
The isolation resistor 40e is a resistor connected to a connection
point 40f and a connection point 40g. The connection point 40f
connects the .pi.-type HPF 40a and the T-type LPF 40b at an output
side of the .pi.-type HPF 40a. The connection point 40g connects
the .pi.-type HPF 40c and the T-type LPF 40d at an output side of
the .pi.-type HPF 40c. The isolation resistor 40e has a resistance
value that depends on a sprit ratio of signal power to the
transmission lines 3 and 4.
The Doherty amplifier of FIG. 7 is designed to equalize electrical
lengths of one path and the other path at a desired band. The one
is a path from the input terminal 1 to an output terminal 12
through the .pi.-type HPF 40a, the T-type LPF 40b, the carrier
amplifier 6, a .lamda./4 line 7, and a power combiner 11. The other
is a path from the input terminal 1 to the output terminal 12
through the .pi.-type HPF 40c, the T-type LPF 40d, the .lamda./4
line 9, a peak amplifier 10, and the power combiner 11.
In the Embodiment 2, for simplifying description, the description
will be given on an assumption that the sprit ratio of the
Wilkinson power divider 40 is 1:1 and input/output impedance of the
Doherty amplifier is 50 .OMEGA..
FIG. 8 is a Smith chart illustrating frequency characteristics of
impedance in the .pi.-type HPFs 40a and 40c and the T-type LPFs 40b
and 40d.
As illustrated with the dotted line in FIG. 8, the frequency
characteristic of impedance in the .pi.-type HPFs 40a and 40c,
which is obtained by viewing from the isolation resistor 40e,
starts from a short point at lower frequencies, goes through an
inductive area, and is minimized in reflection at the desired band.
That is, the frequency characteristic is started from the impedance
of 0 (0 .OMEGA.) on a real axis of the Smith chart, and has the
minimum reflection in the desired band near the impedance of 1.0
(50 .OMEGA.=50 .OMEGA..times.1.0) on the real axis of the Smith
chart.
At higher frequencies, the frequency characteristic approaches
arbitrary impedance which depends on the sprit ratio of the
Wilkinson power divider 40. In the example of FIG. 7, the frequency
characteristic approaches the impedance of 100 .OMEGA..
At this time, in the desired band, the impedance has a
characteristic to shift from higher resistance to lower resistance
along the real axis from lower frequencies to higher frequencies.
In FIG. 8, this characteristic is illustrated with the dotted left
arrow.
Each of the circuit structure of the .pi.-type HPFs 40a and 40c may
be formed by that in FIG. 4A.
However, each circuit structure for the .pi.-type HPFs 40a and 40c
is not limited to that in FIG. 4A as long as the circuit structure
has the frequency characteristic of impedance illustrated with the
dotted line in FIG. 8. Therefore, the number of stages of each of
the .pi.-type HPFs 40a and 40c may be increased or decreased, or
each of the .pi.-type HPFs 40a and 40c may be formed by a
distributed constant line or the like.
As illustrated with a broken line in FIG. 8, at lower frequencies,
the frequency characteristic of impedance in the T-type LPFs 40b
and 40d, which is obtained by viewing from the isolation resistor
40e, starts from arbitrary impedance that depends on the sprit
ratio of the Wilkinson power divider 40, and is minimized in
reflection at the desired band. That is, the frequency
characteristic is started from the impedance of 50 .OMEGA., and has
the minimum reflection in the desired band near the impedance of
1.0 (50 .OMEGA.=50.times.1.0) on the real axis of the Smith
chart.
Thereafter, the frequency characteristic of impedance in the T-type
LPFs 40b and 40d goes through an inductive area as the frequency
becomes higher and eventually approaches an open point where the
impedance is infinite.
At this time, in the desired band, the impedance has a
characteristic to shift from higher resistance to lower resistance
along the real axis from lower frequencies to higher frequencies.
In FIG. 8, this characteristic is illustrated with the broken left
arrow.
The circuit structure of the T-type LPFs 40b and 40d may include an
input terminal 21, an output terminal 22, inductors 26 and 27, and
a capacitor 28, as illustrated in FIG. 3B.
However, each circuit structure for the T-type LPFs 40b and 40d is
not limited to that in FIG. 3B as long as the circuit structure has
the frequency characteristic of impedance illustrated with the
broken line in FIG. 8. Therefore, the number of stages of each of
the T-type LPFs 40b and 40d may be increased or decreased, and each
of T-type LPFs 40b and 40d may be formed by a distributed constant
line or the like.
Next, an operation will be described.
When a radio frequency (RF) signal of the desired band is input
from the input terminal 1 as the signal to be amplified, the
Wilkinson power divider 40 splits the RF signal between the
transmission line 3 and the transmission line 4.
The sprit ratio of the RF signal for the Wilkinson power divider 40
takes an arbitrary value depending on impedance transformations in
the .pi.-type HPFs 40a and 40c and the T-type LPFs 40b and 40d, and
the resistance value of the isolation resistor 40e.
One of the RF signals, which has been split by the Wilkinson power
divider 40 and output to the transmission line 3, is amplified by
the carrier amplifier 6. The other RF signal output to the
transmission line 4 is amplified by the peak amplifier 10.
The RF signal amplified by the carrier amplifier 6 and the other RF
signal amplified by the peak amplifier 10 are combined by the power
combiner 11. A resulting RF signal combined by the power combiner
11 is output through the output terminal 12.
The operation of the Wilkinson power divider 40 will be
specifically described.
FIG. 9 is an explanatory diagram illustrating an isolation
characteristic of the Wilkinson power divider 40.
In FIG. 9, in addition to the isolation characteristic of the
Wilkinson power divider 40 in FIG. 7, an isolation characteristic
of the conventional division circuitry disclosed in Patent
Literature 1 is also depicted for comparison with the Wilkinson
power divider 40 of FIG. 7. In the division circuitry disclosed in
Patent Literature 1, the four .lamda./4 lines are used in place of
the .pi.-type HPFs 40a and 40c and the T-type LPFs 40b and 40d.
In FIG. 9, in order to distinguish those characteristics, the
Wilkinson power divider 40 of FIG. 7 is labeled "present invention"
and the division circuitry disclosed in Patent Literature 1 is
labeled "conventional type".
As illustrated with the dotted lines in FIG. 9, the isolation
characteristic of the conventional-type division circuitry
indicates that isolation between the two output terminals in the
division circuitry becomes high at the vicinity of the desired
band, whereas it becomes low outside the desired band.
In contrast, as illustrated with the solid lines in FIG. 9, the
isolation characteristic of the Wilkinson power divider 40 of FIG.
7 indicates that isolation between the two output terminals in the
Wilkinson power divider 40 becomes high not only in the vicinity of
the desired band but also outside the desired band. The two output
terminals of the Wilkinson power divider 40 correspond to the
output side of the T-type LPF 40b connected to the carrier
amplifier 6 and the output side of the T-type LPF 40d connected to
the .lamda./4 line 9.
The reason why the isolation becomes high even outside the desired
band as described above is that, while the .pi.-type HPFs 40a and
40c function to isolate a frequency band lower than the desired
band, the T-type LPFs 40b and 40d function to isolate a frequency
band higher than the desired band.
FIGS. 10A to 10C are explanatory diagrams illustrating frequency
dependency of a passage characteristic in the Wilkinson power
divider 40 of FIG. 7.
Specifically, FIG. 10A illustrates a passage characteristic of the
Wilkinson power divider 40. FIG. 10B illustrates frequency
characteristics of impedance in a .pi.-type HPF and a .pi.-type
LPF. FIG. 10C illustrates frequency characteristics of impedance in
a .pi.-type HPF and a T-type LPF.
In FIG. 10B, the Wilkinson power divider has a circuit structure in
which an input-side filter is the .pi.-type HPF and an output-side
filter is the .pi.-type LPF. The .pi.-type HPF has a frequency
characteristic of impedance such that, the impedance shifts from
higher resistance to lower resistance along the real axis from
lower frequencies to higher frequencies in the desired band in the
vicinity of impedance of 1.0 (50 .OMEGA.=50.times.1.0) on the real
axis of the Smith chart. In FIG. 10B, this characteristic is
illustrated with the dotted left arrow.
In contrast, the .pi.-type LPF has a frequency characteristic of
impedance, such that the impedance shifts from lower resistance to
higher resistance along the real axis from lower frequencies to
higher frequencies in the desired band. In FIG. 10B, this
characteristic is illustrated with the broken right arrow.
Therefore, the frequency characteristics of impedance in the
.pi.-type HPF and the .pi.-type LPF indicate that the individual
impedances shift in opposite directions along the real axis from
the lower frequencies to higher frequencies in the desired band. As
a result, only a center frequency of the desired band is matched,
and thereby a passband (S21) becomes a narrowband.
The passage characteristic (S21) illustrated with the dotted line
in FIG. 10A represents that of the Wilkinson power divider
illustrated in FIG. 10B. The passage characteristic (S21) indicates
a narrowband characteristic in which only a signal at the center
frequency of the desired band is allowed to pass through without
attenuation.
The Wilkinson power divider 40 of FIG. 7, which is illustrated in
FIG. 10C, has a circuit structure in which an input-side filter is
the .pi.-type HPFs 40a and 40c and an output-side filter is the
T-type LPFs 40b and 40d. The .pi.-type HPFs 40a and 40c have a
frequency characteristic of impedance in which the impedance shifts
from higher resistance to lower resistance along the real axis from
lower frequencies to higher frequencies in the desired band. In
FIG. 10C, this characteristic is illustrated with the dotted left
arrow.
In addition, the T-type LPFs 40b and 40d have a frequency
characteristic of impedance in which the impedance shifts from
lower resistance to higher resistance along the real axis from
higher frequencies to lower frequencies of the desired band. In
FIG. 10C, this characteristic is illustrated with the broken left
arrow.
Therefore, the frequency characteristics of impedance in the
.pi.-type HPFs 40a and 40c and the T-type LPFs 40b and 40d indicate
that the impedance shifts in the same direction along the real axis
from lower frequencies to higher frequencies in the desired band.
As a result, frequencies are broadly matched with each other even
in the vicinity of the center frequency of the desired band, and
thereby the passband (S21) becomes a wide band.
The passage characteristic (S21) illustrated with the solid line in
FIG. 10A represents the passage characteristic of the Wilkinson
power divider 40 of FIG. 7 included in FIG. 10C. The passage
characteristic (S21) has a wide-band passage characteristic which
includes frequencies in the vicinity of the center frequency of the
desired band.
As is clear from the above, according to the Embodiment 2, the
Wilkinson power divider 40 includes: the .pi.-type HPFs 40a and 40c
connected to the input terminal 1; the T-type LPF 40b having one
end connected to an end of the .pi.-type HPF 40a and having another
end connected to the carrier amplifier 6; the T-type LPF 40d having
the one end connected to an end of the .pi.-type HPF 40c and having
another end connected to an end of the .lamda./4 line 9; and the
isolation resistor 40e connected between the connection point 40f
and the connection point 40g. Therefore, it is capable of bringing
advantages that loop oscillation is suppressed without mounting a
stabilizing circuit and a signal is amplified over a wide band.
That is, by providing the Wilkinson power divider 40 achieving high
isolation, loop oscillation can be suppressed not only in the
vicinity of the desired band but also outside the desired band
without mounting a stabilizing circuit. In addition, a signal can
be amplified over a wide band because the .pi.-type HPFs 40a and
40c and the T-type LPFs 40b and 40d have the frequency
characteristics of impedance, in which the impedance shifts in the
same direction along the real axis from the lower frequencies to
higher frequencies in the desired band.
Embodiment 3
In the foregoing Embodiment 1, the Wilkinson power divider 2 has
the circuit structure in which the input-side filter is the
.pi.-type LPFs 2a and 2c and the output-side filter is the T-type
HPFs 2b and 2d. Alternatively, it is possible to adopt a circuit
structure that is formed by switching the input-side filter and the
output-side filter in a Wilkinson power divider 2.
FIG. 11 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 3 of the present invention. In FIG. 11,
the same reference signs as those in FIG. 1 denote the same or
corresponding part, and thus the description thereof will be
omitted.
A Wilkinson power divider 50 is division circuitry that splits a
signal to be amplified, which is input from an input terminal 1,
between the transmission line 3 and the transmission line 4.
The Wilkinson power divider 50 includes T-type HPFs 50a and 50c,
each being a T-type circuit high-pass filter, .pi.-type LPFs 50b
and 50d, each being a .pi.-type circuit low-pass filter, and an
isolation resistor 50e.
The T-type HPF 50a is a first filter having one end connected to
the input terminal 1. The T-type HPF 50a may have a circuit
structure illustrated in FIG. 4B.
The .pi.-type LPF 50b is a second filter having one end connected
to another end of the T-type HPF 50a and having another end
connected to a carrier amplifier 6. The .pi.-type LPF 50b may have
a circuit structure illustrated in FIG. 3A.
The T-type HPF 50c is a third filter having one end connected to
the input terminal 1. The T-type HPF 50c may have a circuit
structure illustrated in FIG. 4B.
The .pi.-type LPF 50d is a fourth filter having one end connected
to another end of the T-type HPF 50c and having another end
connected to a .lamda./4 line 9. The .pi.-type LPF 50d may have a
circuit structure illustrated in FIG. 3A.
The isolation resistor 50e is a resistor connected to a connection
point 50f and a connection point 50g. The connection point 50f
connects the T-type HPF 50a and the .pi.-type LPF 50b at an output
side of the T-type HPF 50a. The connection point 50g connects the
T-type HPF 50c and the .pi.-type LPF 50d at an output side of the
T-type HPF 50c. The isolation resistor 50e has a resistance value
that depends on a sprit ratio of signal power to the transmission
lines 3 and 4.
The frequency characteristic of impedance in the T-type HPFs 50a
and 50c is similar to the frequency characteristic of impedance of
the T-type HPFs 2b and 2d in FIG. 1 for the Embodiment 1.
In addition, the frequency characteristic of impedance in the
.pi.-type LPFs 50b and 50d is similar to the frequency
characteristic of impedance of the .pi.-type LPFs 2a and 2c in FIG.
1 for the Embodiment 1.
Therefore, similarly to the frequency characteristics of impedance
in the i-type LPFs 2a and 2c and the T-type HPFs 2b and 2d, the
frequency characteristics of impedance in the T-type HPFs 50a and
50c and the .pi.-type LPFs 50b and 50d indicate that the impedance
shifts in the same direction along the real axis from lower
frequencies to higher frequencies in the desired band. As a result,
frequencies are broadly matched with each other even in the
vicinity of the center frequency of the desired band, and thereby
the passband (S21) becomes a wide band.
An isolation characteristic of the Wilkinson-type distributor 50 in
FIG. 11 indicates that isolation between two output terminals in
the Wilkinson-type distributor 50 becomes high not only in the
vicinity of the desired band but also outside the desired band,
similarly to the Wilkinson-type distributor 2 of FIG. 1 for the
Embodiment 1. The two output terminals of the Wilkinson-type
distributor 50 correspond to an output side of the .pi.-type LPF
50b connected to the carrier amplifier 6 and an output side of the
.pi.-type LPF 50d connected to the .lamda./4 line 9.
The reason why the isolation becomes high even outside the desired
band as described above is that, the T-type HPFs 50a and 50c
function to isolate frequency band lower than the desired band, and
the .pi.-type LPFs 50b and 50d function to isolate a frequency band
higher than the desired band.
As is clear from the above, according to the Embodiment 3, the
Wilkinson power divider 50 includes: the T-type HPFs 50a and 50c
connected to the input terminal 1; the .pi.-type LPF 50b having one
end connected to an end of the T-type HPF 50a and having another
end connected to the carrier amplifier 6; the .pi.-type LPF 50d
having one end connected to an end of the T-type HPF 50c and having
another end connected to the .lamda./4 line 9; and the isolation
resistor 50e connected to the connection points 50f and 50g.
Therefore, similarly to the foregoing Embodiment 1, it is capable
of bringing advantages that loop oscillation is suppressed without
mounting a stabilizing circuit and a signal is amplified over a
wide band.
Embodiment 4
In the foregoing Embodiment 2, the Wilkinson power divider 40 has
the circuit structure in which the input-side filter is the
.pi.-type HPFs 40a and 40c and the output-side filter is the T-type
LPFs 40b and 40d. Alternatively, it is possible to adopt a circuit
structure that is formed by switching the input-side filter and the
output-side filter in a Wilkinson power divider 40.
FIG. 12 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 4 of the present invention. In FIG. 12,
the same reference signs as those in FIG. 7 denote the same or
corresponding part, and thus the description thereof will be
omitted.
A Wilkinson power divider 60 is division circuitry that splits a
signal to be amplified, which is input from an input terminal 1,
between the transmission line 3 and the transmission line 4.
The Wilkinson power divider 50 includes T T-type LPFs 60a and 60c,
each being a T-type circuit low-pass filter, .pi.-type HPFs 60b and
60d, each being a .pi.-type circuit high-pass filter, and an
isolation resistor 60e.
The T-type LPF 60a is a first filter having one end connected to
the input terminal 1. The T-type LPF 60a may have a circuit
structure illustrated in FIG. 3B.
The .pi.-type HPF 60b is a second filter having one end connected
to another end of the T-type LPF 60a and having another end
connected to a carrier amplifier 6. The .pi.-type HPF 60b may have
a circuit structure illustrated in FIG. 4A.
The T-type LPF 60c is a third filter having one end connected to
the input terminal 1. The T-type LPF 60c may have a circuit
structure illustrated in FIG. 3B.
The .pi.-type HPF 60d is a fourth filter having one end connected
to another end of the T-type LPF 60c and having another end
connected to a .lamda./4 line 9. The .pi.-type HPF 60d may have a
circuit structure illustrated in FIG. 4A.
The isolation resistor 60e is a resistor connected to a connection
point 60f and a connection point 60g. The connection point 60f
connects the T-type LPF 60a and the .pi.-type HPF 60b at an output
side of the T-type LPF 60a. The connection point 60g connects the
T-type LPF 60c and the .pi.-type HPF 60d at an output side of the
T-type LPF 60c. The isolation resistor 60e has a resistance value
that depends on a sprit ratio of signal power to the transmission
lines 3 and 4.
The frequency characteristic of impedance in the T-type LPFs 60a
and 60c is similar to the frequency characteristic of impedance of
the T-type LPFs 40b and 40d in FIG. 7 for the Embodiment 2.
In addition, the frequency characteristic of impedance in the
.pi.-type HPFs 60b and 60d is similar to the frequency
characteristic of impedance of the .pi.-type HPFs 40a and 40c in
FIG. 7 for the Embodiment 2.
Therefore, similarly to the frequency characteristics of impedance
in the .pi.-type HPFs 40a and 40c and the T-type LPFs 40b and 40d,
the frequency characteristics of impedance in the T-type LPFs 60a
and 60c and the .pi.-type HPFs 60b and 60d indicate that the
impedance shifts in the same direction along the real axis from
lower frequencies to higher frequencies in the desired band. As a
result, frequencies are broadly matched with each other even in the
vicinity of the center frequency of the desired band, and thereby
the passband (S21) becomes a wide band.
An isolation characteristic of the Wilkinson-type distributor 60 in
FIG. 12 indicates that isolation between two output terminals in
the Wilkinson-type distributor 60 becomes high not only in the
vicinity of the desired band but also outside the desired band,
similarly to the Wilkinson-type distributor 40 of FIG. 7 for the
Embodiment 2. The two output terminals of the Wilkinson-type
distributor 60 correspond to an output side of the .pi.-type HPF
60b connected to the carrier amplifier 6 and an output side of the
.pi.-type HPF 60d connected to the .lamda./4 line 9.
The reason why the isolation becomes high even outside the desired
band as described above is that, the .pi.-type HPFs 60b and 60d
function to isolate frequency band lower than the desired band, and
the T-type LPFs 60a and 60c function to isolate a frequency band
higher than the desired band.
As is clear from the above, according to the Embodiment 4, the
Wilkinson power divider 60 includes: the T-type LPFs 60a and 60c
connected to the input terminal 1; the .pi.-type HPF 60b having one
end connected to an end of the T-type LPF 60a and having another
end connected to the carrier amplifier 6; the .pi.-type HPF 60d
having one end connected to an end of the T-type LPF 60c and having
another end connected to the .lamda./4 line 9; and the isolation
resistor 60e connected to the connection points 60f and 60g.
Therefore, similarly to the foregoing Embodiment 2, it is capable
of bringing advantages that loop oscillation is suppressed without
mounting a stabilizing circuit and a signal is amplified over a
wide band.
Embodiment 5
In the foregoing Embodiment 1, the amplifier circuitry 8 is formed
with the .lamda./4 line 9 and the peak amplifier 10. Alternatively,
the amplifier circuitry 8 may be formed by introducing a .pi.-type
LPF in place of the .lamda./4 line 9.
FIG. 13 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 5 of the present invention. In FIG. 13,
the same reference sign as that in FIG. 1 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A .pi.-type LPF 71 is a .pi.-type circuit low-pass filter connected
between a T-type HPF 2d and a peak amplifier 10. The .pi.-type LPF
71 has an electrical length being an n-quarter (n is a positive odd
number) wavelength of a signal to be amplified, similarly to the
.lamda./4 line 9 in FIG. 1 for the foregoing Embodiment 1.
Therefore, the .pi.-type LPF 71 serves as a second n-quarter
wavelength line.
Although the circuit structure of the .pi.-type LPF 71 may have a
circuit structure illustrated in FIG. 3A, it is not limited to the
one in FIG. 3A. The number of stages of the .pi.-type LPF 71 may be
increased or decreased, or the .pi.-type LPF 71 may be formed by a
distributed constant line or the like.
The electrical length of the .pi.-type LPF 71 is an n-quarter (n is
a positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 7 of amplifier circuitry 5.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a .pi.-type LPF 2a, a
T-type HPF 2b, a carrier amplifier 6, a .lamda./4 line 7, and a
power combiner 11. The other is a path from the input terminal 1 to
the output terminal 12 via a .pi.-type LPF 2c, the T-type HPF 2d,
the .pi.-type LPF 71, the peak amplifier 10, and the power combiner
11.
The frequency characteristic of impedance in the .pi.-type LPF 71
obtained by viewing from a Wilkinson power divider 2 is similar to
the frequency characteristic of impedance in .pi.-type LPFs 2a and
2c. Therefore, the frequency characteristic indicates starting at
lower frequencies from arbitrary impedance that depends on input
impedance of the peak amplifier 10, and becoming a minimum
reflection at a desired band. At higher frequencies, the frequency
characteristic indicates going through a capacitive area and
approaching a short point as the frequency becomes higher.
In the desired band, the impedance has a characteristic to shift
from lower resistance to higher resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the T-type
HPF 2d and the .pi.-type LPF 71 obtained by viewing from a
connection point of the T-type HPF 2d and the .pi.-type LPF 71 are
similar to those in .pi.-type LPFs and T-type HPFs illustrated in
FIG. 6C. Therefore, the impedance in the desired band shifts in the
same direction along the real axis from lower frequencies to higher
frequencies. As a result, frequencies are widely matched even in
the vicinity of a center frequency of the desired band, and a
passband is widened.
Therefore, by using the .pi.-type LPF 71 for the amplifier
circuitry 8 in place of the .lamda./4 line 9 of FIG. 1, it is
capable of amplifying a signal over a wide band more than the
foregoing Embodiment 1.
Embodiment 6
In the foregoing Embodiment 2, the amplifier circuitry 8 is formed
with the .lamda./4 line 9 and the peak amplifier 10. Alternatively,
the amplifier circuitry 8 may be formed by introducing a .pi.-type
HPF in place of the .lamda./4 line 9.
FIG. 14 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 6 of the present invention. In FIG. 14,
the same reference sign as that in FIG. 7 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A .pi.-type HPF 72 is a .pi.-type circuit high-pass filter
connected between a T-type LPF 40d and a peak amplifier 10. The
.pi.-type HPF 72 has an electrical length being an n-quarter (n is
a positive odd number) wavelength of a signal to be amplified,
similarly to the .lamda./4 line 9 in FIG. 7 for the foregoing
Embodiment 2. Therefore, the .pi.-type HPF 72 serves as a second
n-quarter wavelength line.
Although the circuit structure of the .pi.-type HPF 72 may have a
circuit structure illustrated in FIG. 4A, it is not limited to the
one in FIG. 4A. The number of stages of the .pi.-type HPF 72 may be
increased or decreased, or the .pi.-type HPF 72 may be formed by a
distributed constant line or the like.
The electrical length of the .pi.-type HPF 72 is an n-quarter (n is
a positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 7 of amplifier circuitry 5.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a .pi.-type HPF 40a, a
T-type LPF 40b, a carrier amplifier 6, a .lamda./4 line 7, and a
power combiner 11. The other is a path from the input terminal 1 to
the output terminal 12 via a .pi.-type HPF 40c, the T-type LPF 40d,
the .pi.-type HPF 72, the peak amplifier 10, and the power combiner
11.
The frequency characteristic of impedance in the .pi.-type HPF 72
obtained by viewing from a Wilkinson power divider 40 is similar to
the frequency characteristic of impedance in .pi.-type HPFs 40a and
40c. Therefore, the frequency characteristic indicates starting at
lower frequencies from a short point, going through an inductive
area, and becoming a minimum reflection at a desired band. At
higher frequencies, the frequency characteristic approaches
arbitrary impedance that depends on input impedance of the peak
amplifier 10.
In the desired band, the impedance has a characteristic to shift
from higher resistance to lower resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the T-type
LPF 40d and the .pi.-type HPF 72 obtained by viewing from a
connection point of the T-type LPF 40d and the .pi.-type HPF 72 are
similar to those in .pi.-type HPFs and T-type LPFs illustrated in
FIG. 10C. Therefore, the impedance in the desired band shifts in
the same direction along the real axis from lower frequencies to
higher frequencies. As a result, frequencies are widely matched
even in the vicinity of a center frequency of the desired band, and
a passband is widened.
Therefore, by using the .pi.-type HPF 72 for the amplifier
circuitry 8 in place of the .lamda./4 line 9 of FIG. 7, it is
capable of amplifying a signal over a wide band more than the
foregoing Embodiment 2.
Embodiment 7
In the foregoing Embodiment 3, the amplifier circuitry 8 is formed
with the .lamda./4 line 9 and the peak amplifier 10. Alternatively,
the amplifier circuitry 8 may be formed by introducing a T-type HPF
in place of the .lamda./4 line 9.
FIG. 15 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 7 of the present invention. In FIG. 15,
the same reference sign as that in FIG. 11 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A T-type HPF 73 is a T-type circuit high-pass filter connected
between a i-type LPF 50d and a peak amplifier 10. The T-type HPF 73
has an electrical length being an n-quarter (n is a positive odd
number) wavelength of a signal to be amplified, similarly to the
.lamda./4 line 9 in FIG. 11 for the foregoing Embodiment 3.
Therefore, the T-type HPF 73 serves as a second n-quarter
wavelength line.
Although the circuit structure of the T-type HPF 73 may have a
circuit structure illustrated in FIG. 4B, it is not limited to the
one in FIG. 4B. The number of stages of the T-type HPF 73 may be
increased or decreased, or the T-type HPF 73 may be formed by a
distributed constant line or the like.
The electrical length of the T-type HPF 73 is an n-quarter (n is a
positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 7 of amplifier circuitry 5.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a T-type HPF 50a, a
.pi.-type LPF 50b, a carrier amplifier 6, a .lamda./4 line 7, and a
power combiner 11. The other is a path from the input terminal 1 to
the output terminal 12 via a T-type HPF 50c, the .pi.-type LPF 50d,
the T-type HPF 73, the peak amplifier 10, and the power combiner
11.
The frequency characteristic of impedance in the T-type HPF 73
obtained by viewing from a Wilkinson power divider 50 is similar to
the frequency characteristic of impedance in T-type HPFs 50a and
50c. Therefore, the frequency characteristic indicates starting at
lower frequencies from an open point, going through a capacitive
area, and becoming a minimum reflection at a desired band. At
higher frequencies, the frequency characteristic approaches
arbitrary impedance that depends on input impedance of the peak
amplifier 10.
In the desired band, the impedance has a characteristic to shift
from lower resistance to higher resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the
.pi.-type LPF 50d and the T-type HPF 73 obtained by viewing from a
connection point of the .pi.C-type LPF 50d and the T-type HPF 73
are similar to those in .pi.-type LPFs and T-type HPFs illustrated
in FIG. 6C. Therefore, the impedance in the desired band shifts in
the same direction along the real axis from lower frequencies to
higher frequencies. As a result, frequencies are widely matched
even in the vicinity of a center frequency of the desired band, and
a passband is widened.
Therefore, by using the T-type HPF 73 for the amplifier circuitry 8
in place of the .lamda./4 line 9 of FIG. 11, it is capable of
amplifying a signal over a wide band more than the foregoing
Embodiment 3.
Embodiment 8
In the foregoing Embodiment 4, the amplifier circuitry 8 is formed
with the .lamda./4 line 9 and the peak amplifier 10. Alternatively,
the amplifier circuitry 8 may be formed by introducing a T-type LPF
in place of the .lamda./4 line 9.
FIG. 16 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 8 of the present invention. In FIG. 16,
the same reference sign as that in FIG. 12 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A T-type LPF 74 is a T-type circuit low-pass filter connected
between a .pi.-type HPF 60d and a peak amplifier 10. The T-type LPF
74 has an electrical length being an n-quarter (n is a positive odd
number) wavelength of a signal to be amplified, similarly to the
.lamda./4 line 9 in FIG. 12 for the foregoing Embodiment 4.
Therefore, the T-type LPF 74 serves as a second n-quarter
wavelength line.
Although the circuit structure of the T-type LPF 74 may have a
circuit structure illustrated in FIG. 3B, it is not limited to the
one in FIG. 3B. The number of stages of the T-type LPF 74 may be
increased or decreased, or the T-type LPF 74 may be formed by a
distributed constant line or the like.
The electrical length of the T-type LPF 74 is an n-quarter (n is a
positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 7 of amplifier circuitry 5.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a T-type LPF 60a, a
.pi.-type HPF 60b, a carrier amplifier 6, a .lamda./4 line 7, and a
power combiner 11. The other is a path from the input terminal 1 to
the output terminal 12 via a T-type LPF 60c, the .pi.-type HPF 60d,
the T-type LPF 74, the peak amplifier 10, and the power combiner
11.
The frequency characteristic of impedance in the T-type LPF 74
obtained by viewing from a Wilkinson power divider 60 is similar to
the frequency characteristic of impedance in T-type LPFs 60a and
60c. Therefore, the frequency characteristic indicates starting at
lower frequencies from arbitrary impedance that depends on input
impedance of the peak amplifier 10, and becoming a minimum
reflection at a desired band. At higher frequencies, the frequency
characteristic indicates going through an inductive area as the
frequency becomes higher, and eventually approaching an open point
where the impedance is infinite.
In the desired band, the impedance has a characteristic to shift
from higher resistance to lower resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the
.pi.-type HPF 60d and the T-type LPF 74 obtained by viewing from a
connection point of the .pi.-type HPF 60d and the T-type LPF 74 are
similar to those in .pi.-type HPFs and T-type LPFs illustrated in
FIG. 10C. Therefore, the impedance in the desired band shifts in
the same direction along the real axis from lower frequencies to
higher frequencies. As a result, frequencies are widely matched
even in the vicinity of a center frequency of the desired band, and
a passband is widened.
Therefore, by using the T-type LPF 74 for the amplifier circuitry 8
in place of the .lamda./4 line 9 of FIG. 12, it is capable of
amplifying a signal over a wide band more than the foregoing
Embodiment 4.
Embodiment 9
In the foregoing Embodiments 1 to 4, the example has been
described, in which the .lamda./4 line 7 is connected at a
subsequent stage of the carrier amplifier 6 while the .lamda./4
line 9 is connected at a preceding stage of the peak amplifier 10.
However, as illustrated in FIGS. 17 to 20, an operation can be
performed similarly to the Embodiments 1 to 4, even if a .lamda./4
line 7 is connected at a preceding stage of a carrier amplifier 6
while a .lamda./4 line 9 is connected at a subsequent stage of a
peak amplifier 10.
Accordingly, even a Doherty amplifier, in which the .lamda./4 line
7 is connected at the preceding stage of the carrier amplifier 6
while the .lamda./4 line 9 is connected at the subsequent stage of
the peak amplifier 10, is capable of suppressing loop oscillation
without mounting a stabilizing circuit and also amplifying a signal
over a wide band, similarly to the Embodiments 1 to 4.
Embodiment 10
In the Embodiment 9, the example has been described, in which the
amplifier circuitry 5 connected to the Wilkinson power divider 2 in
the Doherty amplifier of FIG. 17 is formed with the .lamda./4 line
7 and the carrier amplifier 6. Alternatively, amplifier circuitry 5
may be formed by introducing a .pi.-type LPF in place of a
.lamda./4 line 7.
FIG. 21 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 10 of the present invention. In FIG. 21,
the same reference sign as that in FIG. 17 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A .pi.-type LPF 81 is a .pi.-type circuit low-pass filter connected
between a T-type HPF 2b and a carrier amplifier 6. The .pi.-type
LPF 81 has an electrical length being an n-quarter (n is a positive
odd number) wavelength of a signal to be amplified, similarly to
the .lamda./4 line 7 of FIG. 17 for the Embodiment 9. Therefore,
the .pi.-type LPF 81 serves as a first n-quarter wavelength
line.
Although a circuit structure of the .pi.-type LPF 81 may be have a
circuit structure illustrated in FIG. 3A, it is not limited to the
one in FIG. 3A. Therefore, the number of stages of the .pi.-type
LPF 81 may be increased or decreased, or the .pi.-type LPF 81 may
be formed by a distributed constant line or the like.
The electrical length of the .pi.-type LPF 81 is an n-quarter (n is
a positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 9 of amplifier circuitry 8.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a .pi.-type LPF 2a, the
T-type HPF 2b, the .pi.-type LPF 81, the carrier amplifier 6, and a
power combiner 11. The other is a path from the input terminal 1 to
the output terminal 12 via a .pi.-type LPF 2c, a T-type HPF 2d, a
peak amplifier 10, a .lamda./4 line 9, and the power combiner
11.
The frequency characteristic of impedance in the .pi.-type LPF 81
obtained by viewing from a Wilkinson power divider 2 is similar to
the frequency characteristic of impedance in .pi.-type LPFs 2a and
2c. Therefore, the frequency characteristic indicates starting at
lower frequencies from arbitrary impedance that depends on input
impedance of the carrier amplifier 6, and becoming a minimum
reflection at a desired band. At higher frequencies, the frequency
characteristic indicates going through a capacitive area and
approaching a short point as the frequency becomes higher.
In the desired band, the impedance has a characteristic to shift
from lower resistance to higher resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the T-type
HPF 2d and the .pi.-type LPF 81 obtained by viewing from a
connection point of the T-type HPF 2d and the .pi.-type LPF 81 are
similar to those in .pi.-type LPFs and T-type HPFs illustrated in
FIG. 6C. Therefore, the impedance in the desired band shifts in the
same direction along the real axis from lower frequencies to higher
frequencies. As a result, frequencies are widely matched even in
the vicinity of a center frequency of the desired band, and a
passband is widened.
Therefore, by using the .pi.-type LPF 81 for the amplifier
circuitry 5 in place of the .lamda./4 line 7 of FIG. 17, it is
capable of amplifying a signal over a wide band more than the
foregoing Embodiment 9.
Embodiment 11
In the Embodiment 9, the example has been described, in which the
amplifier circuitry 5 connected to the Wilkinson power divider 40
in the Doherty amplifier of FIG. 18 is formed with the .lamda./4
line 7 and the carrier amplifier 6. Alternatively, amplifier
circuitry 5 may be formed by introducing a .pi.-type HPF in place
of a .lamda./4 line 7.
FIG. 22 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 11 of the present invention. In FIG. 22,
the same reference sign as that in FIG. 18 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A .pi.-type HPF 82 is a .pi.-type circuit high-pass filter
connected between a T-type LPF 40b and a carrier amplifier 6. The
.pi.-type HPF 82 has an electrical length being an n-quarter (n is
a positive odd number) wavelength of a signal to be amplified,
similarly to the .lamda./4 line 7 in FIG. 18 for the foregoing
Embodiment 9. Therefore, the .pi.-type HPF 82 serves as a first
n-quarter wavelength line.
Although the circuit structure of the .pi.-type HPF 82 may have a
circuit structure illustrated in FIG. 4A, it is not limited to the
one in FIG. 4A. The number of stages of the .pi.-type HPF 82 may be
increased or decreased, or the .pi.-type HPF 82 may be formed by a
distributed constant line or the like.
The electrical length of the .pi.-type HPF 82 is an n-quarter (n is
a positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 8 of amplifier circuitry 8.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a .pi.-type HPF 40a, the
T-type LPF 40b, the .pi.-type HPF 82, the carrier amplifier 6, and
a power combiner 11. The other is a path from the input terminal 1
to the output terminal 12 via a .pi.-type HPF 40c, a T-type LPF
40d, a peak amplifier 10, a .lamda./4 line 9, and the power
combiner 11.
The frequency characteristic of impedance in the .pi.-type HPF 82
obtained by viewing from a Wilkinson power divider 40 is similar to
the frequency characteristic of impedance in .pi.-type HPFs 40a and
40c. Therefore, the frequency characteristic indicates starting at
lower frequencies from a short point, going through an inductive
area, and becoming a minimum reflection at a desired band. At
higher frequencies, the frequency characteristic approaches
arbitrary impedance that depends on input impedance of the carrier
amplifier 6.
In the desired band, the impedance has a characteristic to shift
from higher resistance to lower resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the T-type
LPF 40b and the .pi.-type HPF 82 obtained by viewing from a
connection point of the T-type LPF 40b and the .pi.-type HPF 82 are
similar to those in .pi.-type HPFs and T-type LPFs illustrated in
FIG. 10C. Therefore, the impedance in the desired band shifts in
the same direction along the real axis from lower frequencies to
higher frequencies. As a result, frequencies are widely matched
even in the vicinity of a center frequency of the desired band, and
a passband is widened.
Therefore, by using the .pi.-type HPF 82 for the amplifier
circuitry 5 in place of the .lamda./4 line 7 of FIG. 18, it is
capable of amplifying a signal over a wide band more than the
foregoing Embodiment 9.
Embodiment 12
In the foregoing Embodiment 9, the example has been described, in
which the amplifier circuitry 5 connected to the Wilkinson power
divider 50 in the Doherty amplifier of FIG. 19 is formed with the
.lamda./4 line 7 and the carrier amplifier 6. Alternatively,
amplifier circuitry 5 may be formed by introducing a T-type HPF in
place of a .lamda./4 line 7.
FIG. 23 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 12 of the present invention. In FIG. 23,
the same reference sign as that in FIG. 19 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A T-type HPF 83 is a T-type circuit high-pass filter connected
between a .pi.-type LPF 50b and a carrier amplifier 6. The T-type
HPF 83 has an electrical length being an n-quarter (n is a positive
odd number) wavelength of a signal to be amplified, similarly to
the .lamda./4 line 7 in FIG. 19 for the foregoing Embodiment 9.
Therefore, the T-type HPF 83 serves as a first n-quarter wavelength
line.
Although the circuit structure of the T-type HPF 83 may have a
circuit structure illustrated in FIG. 4B, it is not limited to the
one in FIG. 4B. The number of stages of the T-type HPF 83 may be
increased or decreased, or the T-type HPF 83 may be formed by a
distributed constant line or the like.
The electrical length of the T-type HPF 83 is an n-quarter (n is a
positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 9 of amplifier circuitry 8.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a T-type HPF 50a, the
.pi.-type LPF 50b, the T-type HPF 83, the carrier amplifier 6, and
a power combiner 11. The other is a path from the input terminal 1
to the output terminal 12 via a T-type HPF 50c, a .pi.-type LPF
50d, a peak amplifier 10, a .lamda./4 line 9, and the power
combiner 11.
The frequency characteristic of impedance in the T-type HPF 83
obtained by viewing from a Wilkinson power divider 50 is similar to
the frequency characteristic of impedance in T-type HPFs 50a and
50c. Therefore, the frequency characteristic indicates starting at
lower frequencies from an open point, going through a capacitive
area, and becoming a minimum reflection at a desired band. At
higher frequencies, the frequency characteristic approaches
arbitrary impedance that depends on input impedance of the carrier
amplifier 6.
In the desired band, the impedance has a characteristic to shift
from lower resistance to higher resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the
.pi.-type LPF 50b and the T-type HPF 83 obtained by viewing from a
connection point of the .pi.-type LPF 50b and the T-type HPF 83 are
similar to those in .pi.-type LPFs and T-type HPFs illustrated in
FIG. 6C. Therefore, the impedance in the desired band shifts in the
same direction along the real axis from lower frequencies to higher
frequencies. As a result, frequencies are widely matched even in
the vicinity of a center frequency of the desired band, and a
passband is widened.
Therefore, by using the T-type HPF 83 for the amplifier circuitry 8
in place of the .lamda./4 line 7 of FIG. 19, it is capable of
amplifying a signal over a wide band more than the foregoing
Embodiment 9.
Embodiment 13
In the Embodiment 9, the example has been described, in which the
amplifier circuitry 5 connected to the Wilkinson power divider 60
in the Doherty amplifier of FIG. 20 is formed with the .lamda./4
line 7 and the carrier amplifier 6. Alternatively, amplifier
circuitry 5 may be formed by introducing a T-type LPF in place of a
.lamda./4 line 7.
FIG. 24 is a configuration diagram illustrating a Doherty amplifier
according to an Embodiment 13 of the present invention. In FIG. 24,
the same reference sign as that in FIG. 20 denotes the same or
corresponding part, and thus the description thereof will be
omitted.
A T-type LPF 84 is a T-type circuit low-pass filter connected
between a .pi.-type HPF 60b and a carrier amplifier 6. The T-type
LPF 84 has an electrical length being an n-quarter (n is a positive
odd number) wavelength of a signal to be amplified, similarly to
the .lamda./4 line 7 in FIG. 20 for the foregoing Embodiment 9.
Therefore, the T-type LPF 84 serves as a first n-quarter wavelength
line.
Although the circuit structure of the T-type LPF 84 may have a
circuit structure illustrated in FIG. 3B, it is not limited to the
one in FIG. 3B. The number of stages of the T-type LPF 84 may be
increased or decreased, or the T-type LPF 84 may be formed by a
distributed constant line or the like.
The electrical length of the T-type LPF 84 is an n-quarter (n is a
positive odd number) wavelength of the signal to be amplified,
similarly to a .lamda./4 line 9 of amplifier circuitry 8.
Therefore, at a desired band, electrical lengths are equal between
one path and the other path. The one is a path from an input
terminal 1 to an output terminal 12 via a T-type LPF 60a, the
.pi.-type HPF 60b, the T-type LPF 84, the carrier amplifier 6, and
a power combiner 11. The other is a path from the input terminal 1
to the output terminal 12 via a T-type LPF 60c, a .pi.-type HPF
60d, a peak amplifier 10, a .lamda./4 line 9, and the power
combiner 11.
The frequency characteristic of impedance in the T-type LPF 84
obtained by viewing from a Wilkinson power divider 60 is similar to
the frequency characteristic of impedance in T-type LPFs 60a and
60c. Therefore, the frequency characteristic indicates starting at
lower frequencies from arbitrary impedance that depends on input
impedance of the carrier amplifier 6, and becoming a minimum
reflection at a desired band. At higher frequencies, the frequency
characteristic indicates going through an inductive area as the
frequency becomes higher, and eventually approaching an open point
where the impedance is infinite.
In the desired band, the impedance has a characteristic to shift
from higher resistance to lower resistance along a real axis from
lower frequencies to higher frequencies.
The individual frequency characteristics of impedance of the
.pi.-type HPF 60b and the T-type LPF 84 obtained by viewing from a
connection point of the .pi.-type HPF 60b and the T-type LPF 84 are
similar to those in .pi.-type HPFs and T-type LPFs illustrated in
FIG. 10C. Therefore, the impedance in the desired band shifts in
the same direction along the real axis from lower frequencies to
higher frequencies. As a result, frequencies are widely matched
even in the vicinity of a center frequency of the desired band, and
a passband is widened.
Therefore, by using the T-type LPF 84 for the amplifier circuitry 5
in place of the .lamda./4 line 7 of FIG. 20, it is capable of
amplifying a signal over a wide band more than the foregoing
Embodiment 9.
Note that the invention of the present application allows free
combinations of the embodiments, modifications of arbitrary
configuration elements of the embodiments, or omissions of
arbitrary configuration elements in the embodiments, within the
scope of the invention.
The Doherty amplifier according to the present invention is
suitable for an amplifier that needs to amplify a signal over a
wide band.
REFERENCE SIGNS LIST
1: Input terminal, 2: Wilkinson power divider (division circuitry),
2a: .pi.-type LPF (first filter), 2b: T-type HPF (second filter),
2c: it-type LPF (third filter), 2d: T-type HPF (fourth filter), 2e:
Isolation resistor (resistor), 2f and 2g: Connection point, 3:
Transmission line (first transmission line), 4: Transmission line
(second transmission line), 5: Amplifier circuitry (first amplifier
circuitry), 6: Carrier amplifier, 7: .lamda./4 line (first
n-quarter wavelength line), 8: Amplifier circuitry (second
amplifier circuitry), 9: .lamda./4 line (second n-quarter
wavelength line), 10: Peak amplifier, 11: Power combiner, 12:
Output terminal, 21: Input terminal, 22: Output terminal, 23 and
24: Capacitor, 25: Inductor, 26 and 27: Inductor, 28: Capacitor,
31: Input terminal, 32: Output terminal, 33 and 34: Capacitor, 35:
Inductor, 36 and 37: Capacitor, 38: Inductor, 40: Wilkinson power
divider, 40a: it-type HPF (first filter), 40b: T-type LPF (second
filter), 40c: .pi.-type HPF (third filter), 40d: T-type LPF (fourth
filter), 40e: Isolation resistor (resistor), 40f and 40g:
Connection point, 50: Wilkinson power divider, 50a: T-type HPF
(first filter), 50b: .pi.-type LPF (second filter), 50c: T-type HPF
(third filter), 50d: .pi.-type LPF (fourth filter), 50e: Isolation
resistor (resistor), 50f and 50g: Connection point, 60: Wilkinson
power divider, 60a: T-type LPF (first filter), 60b: .pi.-type HPF
(second filter), 60c: T-type LPF (third filter), 60d: .pi.-type HPF
(fourth filter), 60f and 60g: Connection point, 71: .pi.-type LPF
(second n-quarter wavelength line), 72: .pi.-type HPF (second
n-quarter wavelength line), 73: T-type HPF (second n-quarter
wavelength line), 74: T-type LPF (second n-quarter wavelength
line), 81: .pi.-type LPF (first n-quarter wavelength line), 82:
.pi.-type HPF (first n-quarter wavelength line), 83: T-type HPF
(first n-quarter wavelength line), 84: T-type LPF (first n-quarter
wavelength line)
* * * * *